All experiments have been carried out under the University of California, San Diego's Animal Care and Use Committee guidelines.
For shRNA-mediated knockdown, candidate shRNA sequences were purchased from Open Biosystems (Lafayette, CO, USA) in the pLKO.1 expression vector driving shRNA expression off the U6 promoter and evaluated for efficacy and specificity. For shRNA-mediated knockdown of NeuroD2, we used the following sequence corresponding to nucleotides 911 to 931 of rat NeuroD2: GCTCTGTCTCAACGGCAACTT. Mouse and rat NeuroD2 are 100% conserved in the target region for the shRNA. For in vivo knockdown of NeuroD2 expression, the U6 promoter and shNeuroD2 sequence from PLKO.1 were cloned into the PacI site of the lentiviral vector plasmid FCK(0.4)GW (a gift from Dr Pavel Osten, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), which contains a 0.4 kb fragment of the mouse CaMKII promoter driving EGFP . As a control vector, we used FCK(0.4)GW alone. Myc-tagged mouse NeuroD2 and NeuroD1 were expressed from the pCS2+ plasmid off the CMV promoter. Validated lentiviral constructs expressing shRNA against PSD95 and empty vector control expressing GFP alone were obtained in the pLLox3.7 vector (a gift from Dr Roger Nicoll, UCSF, San Francisco, CA). The shRNA sequence for PSD95 was: TCACGATCATCGCTCAGTATA. These constructs have been extensively validated for efficacy and specificity in previous publications . A second independent PSD95 shRNA and GFP expressing lentiviral construct in the pGIPZ vector (Open Biosystems) was used for confirmation, for which the sequence was CAGCACATCCCTGGAGATA. This second construct or a control construct (FCK(0.4)GFP) was combined with pFCK(0.4)mGFP, expressing membrane-targeted GFP for in utero electroporation experiments.
Intracellular injection of Lucifer Yellow
Mice of either sex were anesthetized and transcardially perfused with phosphate-buffered saline (PBS) followed by 4% paraformaldehyde at pH 7.4 and post-fixed in 4% paraformaldehyde for 1 hour on ice. Then, 100 μm thick coronal sections were cut using a vibratome and stored in PBS on ice. Penetrating microelectrodes were pulled from standard borosilicate capillary glass with filament (1 mm outer diameter/0.58 mm inner diameter) and back-filled with LY dye (5%). Slices were mounted on coverslips under PBS and CA3 neurons were filled via iontophoresis under visual guidance. Sections were then post-fixed an additional 15 minutes before being prepared for immunohistochemistry.
In utero electroporation
Timed-pregnant CD-1 white mice (Charles River, E15) were anesthetized with 3% isoflurane. A small vertical incision was made in the skin and abdominal wall and embryos gently exposed. Each embryo was injected with 1 to 2 μl of DNA solution and 0.01% Fast Green into the ventricles . For spine analysis, pGIPZ-shPSD95 plasmid DNA or pFCK(0.4) was mixed with pFCK(0.4)-mGFP DNA. A pressure-controlled beveled glass pipette (Drummond (Broomall, PA, USA), WPI Microbeveller (Sarasota, FL, USA)) was used for injection. After each injection, the embryos were moistened with PBS and voltage steps via tweezertrodes (BTX, 5 mm round, platinum, BTX electroporator (Holliston, MA, USA)) were applied at a 30 to 45 degree angle with respect to the interaural line to target CA3. Voltage was 36V for 5 pulses at 1 Hz, each pulse lasting 50 ms.
Imaging and analysis of spines
Spines were imaged on a Leica SP2 or SP5 confocal microscope under 63 × magnification with 3 × optical zoom for imaging of TE spines and 4 × optical zoom for imaging of classical spines. Images were collected from clearly identifiable pyramidal neurons throughout the extent of CA3 with TE spine images collected on primary and secondary dendrites and classic spine images collected on tertiary branches in the middle third of the CA3 stratum radiatum. Images were collected as z-stacks with 0.5 μm thick sections. Images were analyzed as confocal stacks using ImageJ software. For all spines, individual spine heads were identified and analyzed in the confocal plane in which they had the largest area. Head width measurements were obtained using a custom ImageJ plugin called edgefitter . Individual multi-headed spines cannot be separated at the light level, but individual spine heads could be reliably separated using high-resolution confocal stacks and our analysis is of overall TE spine head density.
Lentivirus was made by transfecting 293T cells with the pFCK(0.4)GW plasmid along with helper plasmids (psPAX2 and VSVG). After 3 days, 293T media was centrifuged (46,000 × g) to concentrate the virus, resuspended in PBS, and stored at -80°C. For in vivo injections, P5 rat pups of either sex were anesthetized using an isoflurane vaporizer and immobilized in a stereotaxic device. Following craniotomy, a Hamilton syringe was used to inject 1 μl of concentrated virus. The animals were sutured and returned to their cage until further analysis.
Whole-cell, voltage-clamp recordings were performed on CA3 pyramidal neurons in acute brain slices from NeuroD2 null mice and littermate controls. Mice of either sex aged P14 to P17 were deeply anesthetized with isoflurane and then rapidly decapitated. Brain slices were cut in the sagittal plane at a thickness of 350 μm using a vibrating microtome (VT-1200, Leica Microsystems, Bannockburn, IL, USA). Slicing was performed in an ice-cold artificial cerebral spinal fluid with the following ionic composition (concentrations in mM): NaCl 124, KCl 5, NaHCO3 26, NaH2PO4 1.25, glucose 10, MgCl2 6, CaCl2 1. Slices were moved directly to a holding chamber and maintained for at least 0.5 hours (0.5 to 3 hours), and then transferred to a recording chamber. In both the holding and recording chambers, slices were submerged in a standard artificial cerebral spinal fluid solution (at room temperature) with the following ionic composition (in mM): NaCl 124, KCl 5, NaHCO3 26, NaH2PO4 1.25, glucose 10, MgCl2 3, CaCl2 2, and bubbled constantly with 95% O2:5% CO2 gas. Our recording solution also contained 100 μM picrotoxin (Tocris Bioscience (Minneapolis, MN, USA)) to block GABA receptor-mediated inhibitory currents. CA3 pyramidal neurons were visualized by infrared differential interference contrast imaging using an upright, fixed-stage microscope (BX-51, Olympus (Center Valley, PA, USA)). Whole-cell patch electrodes were pulled from borosilicate glass (1.5 mm outer diameter and 1.16 mm inner diameter, Warner Instruments (Hamden, CT, USA)) to resistances ranging from 3 to 6 MΩ. Access resistance was monitored for consistency during recordings and ranged from 10 to 20 MΩ for the cells included for analysis. The intracellular recording solution contained (in mM): CsCl 20, CsMeSO3 105, ATP (dipotassium salt) 0.5, GTP 0.3, Hepes 10, MgCl2 2, EGTA 1, QX-314 2 and BAPTA 10, at pH 7.3. Recordings were acquired using a PC-505 amplifier (Warner Instruments (Hamden, CT, USA)) and digitized using custom software routines written in Igor Pro (Wavemetrics (Portland, OR, USA)) and the NIDAQ tools package, to access a PCI-based board (PCI-1200, National Instruments (Austin, TX, USA)) on a Macintosh G3 computer.
Hippocampal neurons were cultured from P0 WT and NeuroD2 null mouse littermates of either sex. The whole hippocampus, including DG, CA3 and CA1 regions, was dissected free from the cortex, and neurons were dissociated and plated on a rat glial monolayer previously cultured on poly-D-lysine (Millipore, Temecula, CA, USA) and laminin (Invitrogen, Carlsbad, CA, USA) coated coverslips. Neurons were maintained in Neurobasal-A medium (Invitrogen) supplemented with B27, glucose, glutamax, penicillin/streptomycin (Invitrogen) and 25 μM B-mercaptoethanol.
For LY immunohistochemistry, slices were blocked 1 hour at room temperature in 3% bovine serum albumin + 0.3% Triton X-100 in PBS and incubated in primary antibody against LY (Abcam, rabbit anti-LY (Cambridge, MA, USA); 1:1,000) overnight at 4°C in blocking solution. Slices were then washed three times for 20 minutes each with blocking solution and incubated in secondary antibody (Molecular Probes, donkey anti-rabbit 555 (Grand Island, NY, USA); 1:1,000) for 2 hours at room temperature. Finally, slices were labeled with a Hoechst nuclear stain and mounted on slides for confocal microscopy. For virally labeled neurons, slices were simultaneously incubated with primary antibody against GFP (Abcam, goat anti-GFP; 1:3,000) and stained with a secondary antibody in the far red channel (Molecular Probes, donkey anti-goat 647, 1:1,000).
For immunohistochemistry in slices from WT and NeuroD2 null mice, animals of either sex were transcardially perfused as for microinjection of LY, but were post-fixed overnight at 4°C. Slices for immunohistochemistry were saturated with a 30% sucrose in PBS solution for approximately 3 days and flash frozen in dry ice before cutting 30 μm frozen sections on a cryostat microtome. Slices were then blocked with 3% bovine serum albumin in PBS for 1 hour before being incubated in primary antibody overnight at 4°C, washed and incubated with fluorophore conjugated secondary antibody for 2 hours at room temperature, and washed and incubated with Hoechst nuclear stain before being coverslipped.
Cultured neurons were fixed in 4% paraformaldehyde, 4% sucrose in PBS and processed for immunofluorescence according to standard procedures. Cells were washed with PBS and incubated in blocking solution (PBS plus 3% bovine albumin and 0.1% TritonX 100) for 30 minutes. For SPO staining only, 0.1% Saponin was included only in the preliminary blocking step. Then cells were incubated in primary antibody diluted in blocking solution for 2 hours, washed, incubated in secondary antibody diluted in blocking solution for 1 hour, washed again and stained with Hoechst nuclear stain before being coverslipped. Primary antibodies were: rat anti-CTIP2 1:1,000 (Abcam), rabbit anti-synaptoporin 1:1,000 (Synaptic Systems, (Goettingen, Germany)), mouse anti-PSD95 1:250 (NeuroMAB, (Davis, CA, USA)), guinea-pig anti-Vglut1 1:5,000 (Chemicon, (Billerica, MA, USA)), Chicken anti-MAP2 1:5,000 (Abcam). Fluorophore-conjugated secondary antibodies were from Jackson ImmunoResearch (West Grove, PA, USA) or Invitrogen and were used at 1:1,000.
Image acquisition and analysis of in vitro experiments
Images were captured on Leica SP2 and SP5 confocal microscopes (Leica Microsystems (Buffalo Grove, IL, USA)). Z-Stacks were collapsed in a maximum projection and analyzed using NIH ImageJ software. Images were thresholdeded using constant settings per experiment and colocalized synaptic markers were quantified per unit length of dendrite using an automated method of assessing pixel overlap.
Immunoblotting and quantification
Western blotting was performed by standard procedures. For hippocampal lysates, mice of either sex were decapitated under isoflurane anesthesia, and the hippocampus was dissected free in cold PBS and homogenized in buffer with protease inhibitors. For 293T expression experiments cells were lysed directly in hot SDS containing sample buffer and boiled for 10 minutes. Lysates were diluted into SDS containing sample buffer, boiled for 10 minutes and run by SDS-PAGE. Antibodies used for immunoblotting were the following: mouse anti-myc, 1:1,000 (Santa Cruz Biotechnology (Santa Cruz, CA, USA)), mouse anti-PSD95 1:1,000 (Affinity Bioreagents ABR (Rockford, IL, USA)), mouse anti-βIIItubulin 1:1,000 (Abcam), mouse anti-GAPDH 1:5,000 (Millipore (Billerica, MA, USA)), mouse anti-Pick1 1:500 (NeuroMAB), mouse anti-SAP102 1:1,000 (NeuroMAB), rabbit anti-GluR2 1:1,000 (Chemicon), mouse anti-NR2A 1:1,000 (Chemicon), mouse anti-NR2B 1:1,000 (NeuroMAB), rat anti-synaptoporin 1:1,000 (Synaptic Systems), rabbit anti-CKII 1:1,000 (Upstate Biotechnology (Billerica, MA, USA)), rabbit anti-BDNF 1:1,000 (Santa Cruz Biotechnology). Immunoblotting and quantification for SynCAM1 to -3 were performed as previously described . Quantification of other immunoblots was by measurement of densitometry in standard sized windows, using imageJ software.
Quantitative PCR experiments
Mouse cortical neurons were taken from embryonic 15 day pups of either sex, cultured in Neural Basal Medium with B27 and stimulated with indicated methods at 3 and 12 days in vitro, respectively. Total RNA was collected 3 hours after stimulation and reverse-transcribed to cDNA. Real-time PCR are performed with SYBR-Green dye on Rotor-Gene Q (Qiagen (Valencia, CA, USA)). Statistical analysis is done with software with Rotor-Gene.